[AutoSchedule] Support multiple cache read and fix bugs #6686
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jcf94
approved these changes
Oct 15, 2020
|
The failed case in CI seems related to #6417 |
comaniac
force-pushed
the
ansor_for_backward
branch
from
4ac105d to
e2a7562
Compare
comaniac
force-pushed
the
ansor_for_backward
branch
from
e2a7562 to
b4798c9
Compare
merrymercy
approved these changes
Oct 18, 2020
merrymercy
reviewed
Oct 18, 2020
trevor-m
pushed a commit
to trevor-m/tvm
that referenced
this pull request
Oct 29, 2020
* Add shape to DAG print * avoid useless cross-thread reduction * Fix stage order * support multiple cache_read * lint * fix * fix * address comment * fix ci * Trigger CI & Update doc strings Co-authored-by: Lianmin Zheng <[email protected]>
trevor-m
pushed a commit
to trevor-m/tvm
that referenced
this pull request
Dec 2, 2020
* Add shape to DAG print * avoid useless cross-thread reduction * Fix stage order * support multiple cache_read * lint * fix * fix * address comment * fix ci * Trigger CI & Update doc strings Co-authored-by: Lianmin Zheng <[email protected]>
trevor-m
pushed a commit
to trevor-m/tvm
that referenced
this pull request
Dec 4, 2020
* Add shape to DAG print * avoid useless cross-thread reduction * Fix stage order * support multiple cache_read * lint * fix * fix * address comment * fix ci * Trigger CI & Update doc strings Co-authored-by: Lianmin Zheng <[email protected]>
trevor-m
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Dec 4, 2020
* Add shape to DAG print * avoid useless cross-thread reduction * Fix stage order * support multiple cache_read * lint * fix * fix * address comment * fix ci * Trigger CI & Update doc strings Co-authored-by: Lianmin Zheng <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR improves the cache read sketch generation rule in auto-scheduler to support multiple cache reads. Previously, cache read sketch generation rule will simply give up if a tensor has multiple consumers. This results in all consumers read that tensor directly from the host memory, which hurts the performance a lot. This PR resolves this limitation.
In addition, this PR also fixes the following bugs/issues:
max_threads_per_blockin cross thread reduction rule condition, so that we can guarantee computation intensive ops won't have cross thread reduction sketch, which usually doesn't help..dsuffix to the name of redundant cache read ops to avoid conflict. Without this change, auto-scheduler outputs Python schedule APIs won't work for multiple cache reads.ComputeDAGby taking a schedule, so that we can enforce stage order consistency.cc @merrymercy @jcf94 @tqchen